| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 21900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.091324200913242, | |
| "grad_norm": 4.222515106201172, | |
| "learning_rate": 2.9315068493150685e-05, | |
| "loss": 0.5886, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.182648401826484, | |
| "grad_norm": 4.046519756317139, | |
| "learning_rate": 2.863013698630137e-05, | |
| "loss": 0.5361, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.273972602739726, | |
| "grad_norm": 4.622340679168701, | |
| "learning_rate": 2.7945205479452054e-05, | |
| "loss": 0.53, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.365296803652968, | |
| "grad_norm": 6.696527481079102, | |
| "learning_rate": 2.726027397260274e-05, | |
| "loss": 0.522, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.45662100456621, | |
| "grad_norm": 4.05750846862793, | |
| "learning_rate": 2.6575342465753424e-05, | |
| "loss": 0.523, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.547945205479452, | |
| "grad_norm": 5.323693752288818, | |
| "learning_rate": 2.589041095890411e-05, | |
| "loss": 0.5322, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.639269406392694, | |
| "grad_norm": 5.546619892120361, | |
| "learning_rate": 2.5205479452054793e-05, | |
| "loss": 0.5302, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.730593607305936, | |
| "grad_norm": 4.484456539154053, | |
| "learning_rate": 2.452054794520548e-05, | |
| "loss": 0.5118, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.821917808219178, | |
| "grad_norm": 4.336514949798584, | |
| "learning_rate": 2.3835616438356165e-05, | |
| "loss": 0.5204, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.91324200913242, | |
| "grad_norm": 3.2517621517181396, | |
| "learning_rate": 2.315068493150685e-05, | |
| "loss": 0.518, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5103840827941895, | |
| "eval_runtime": 48.2013, | |
| "eval_samples_per_second": 219.289, | |
| "eval_steps_per_second": 54.833, | |
| "step": 5475 | |
| }, | |
| { | |
| "epoch": 1.004566210045662, | |
| "grad_norm": 4.530373573303223, | |
| "learning_rate": 2.2465753424657534e-05, | |
| "loss": 0.5034, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.095890410958904, | |
| "grad_norm": 3.625715732574463, | |
| "learning_rate": 2.178082191780822e-05, | |
| "loss": 0.4662, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.187214611872146, | |
| "grad_norm": 4.742297649383545, | |
| "learning_rate": 2.1095890410958904e-05, | |
| "loss": 0.4676, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.278538812785388, | |
| "grad_norm": 3.3340749740600586, | |
| "learning_rate": 2.041095890410959e-05, | |
| "loss": 0.4885, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.36986301369863, | |
| "grad_norm": 4.077823638916016, | |
| "learning_rate": 1.9726027397260273e-05, | |
| "loss": 0.4562, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.461187214611872, | |
| "grad_norm": 2.796003580093384, | |
| "learning_rate": 1.904109589041096e-05, | |
| "loss": 0.4597, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5525114155251143, | |
| "grad_norm": 2.7733001708984375, | |
| "learning_rate": 1.8356164383561642e-05, | |
| "loss": 0.4552, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.643835616438356, | |
| "grad_norm": 2.2728240489959717, | |
| "learning_rate": 1.767123287671233e-05, | |
| "loss": 0.4584, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7351598173515983, | |
| "grad_norm": 2.6873130798339844, | |
| "learning_rate": 1.6986301369863014e-05, | |
| "loss": 0.465, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.82648401826484, | |
| "grad_norm": 4.094951152801514, | |
| "learning_rate": 1.63013698630137e-05, | |
| "loss": 0.4704, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9178082191780823, | |
| "grad_norm": 4.901790618896484, | |
| "learning_rate": 1.5616438356164384e-05, | |
| "loss": 0.4636, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5072851181030273, | |
| "eval_runtime": 47.2386, | |
| "eval_samples_per_second": 223.758, | |
| "eval_steps_per_second": 55.95, | |
| "step": 10950 | |
| }, | |
| { | |
| "epoch": 2.009132420091324, | |
| "grad_norm": 3.074185371398926, | |
| "learning_rate": 1.4931506849315068e-05, | |
| "loss": 0.4636, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.1004566210045663, | |
| "grad_norm": 4.926590442657471, | |
| "learning_rate": 1.4246575342465753e-05, | |
| "loss": 0.4476, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.191780821917808, | |
| "grad_norm": 9.777703285217285, | |
| "learning_rate": 1.3561643835616437e-05, | |
| "loss": 0.4311, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.2831050228310503, | |
| "grad_norm": 2.8424761295318604, | |
| "learning_rate": 1.2876712328767124e-05, | |
| "loss": 0.4331, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.374429223744292, | |
| "grad_norm": 7.403332710266113, | |
| "learning_rate": 1.2191780821917808e-05, | |
| "loss": 0.4337, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4657534246575343, | |
| "grad_norm": 3.8425567150115967, | |
| "learning_rate": 1.1506849315068493e-05, | |
| "loss": 0.4315, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.557077625570776, | |
| "grad_norm": 5.38580322265625, | |
| "learning_rate": 1.0821917808219177e-05, | |
| "loss": 0.4344, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.6484018264840183, | |
| "grad_norm": 5.385922908782959, | |
| "learning_rate": 1.0136986301369862e-05, | |
| "loss": 0.4331, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.73972602739726, | |
| "grad_norm": 3.808145523071289, | |
| "learning_rate": 9.452054794520548e-06, | |
| "loss": 0.4476, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.8310502283105023, | |
| "grad_norm": 2.4701247215270996, | |
| "learning_rate": 8.767123287671233e-06, | |
| "loss": 0.4142, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.922374429223744, | |
| "grad_norm": 3.8906421661376953, | |
| "learning_rate": 8.082191780821917e-06, | |
| "loss": 0.412, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.5061534643173218, | |
| "eval_runtime": 46.16, | |
| "eval_samples_per_second": 228.986, | |
| "eval_steps_per_second": 57.257, | |
| "step": 16425 | |
| }, | |
| { | |
| "epoch": 3.0136986301369864, | |
| "grad_norm": 3.2641985416412354, | |
| "learning_rate": 7.397260273972603e-06, | |
| "loss": 0.435, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.105022831050228, | |
| "grad_norm": 3.6989686489105225, | |
| "learning_rate": 6.712328767123287e-06, | |
| "loss": 0.418, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.1963470319634704, | |
| "grad_norm": 3.902902603149414, | |
| "learning_rate": 6.027397260273973e-06, | |
| "loss": 0.4088, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.287671232876712, | |
| "grad_norm": 2.8096017837524414, | |
| "learning_rate": 5.342465753424657e-06, | |
| "loss": 0.4047, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.3789954337899544, | |
| "grad_norm": 4.373694896697998, | |
| "learning_rate": 4.657534246575343e-06, | |
| "loss": 0.4149, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.470319634703196, | |
| "grad_norm": 3.5170211791992188, | |
| "learning_rate": 3.972602739726027e-06, | |
| "loss": 0.406, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.5616438356164384, | |
| "grad_norm": 5.333255290985107, | |
| "learning_rate": 3.2876712328767123e-06, | |
| "loss": 0.4168, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.65296803652968, | |
| "grad_norm": 4.406794548034668, | |
| "learning_rate": 2.6027397260273973e-06, | |
| "loss": 0.4132, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.7442922374429224, | |
| "grad_norm": 3.849666118621826, | |
| "learning_rate": 1.9178082191780823e-06, | |
| "loss": 0.4175, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.8356164383561646, | |
| "grad_norm": 3.263227701187134, | |
| "learning_rate": 1.232876712328767e-06, | |
| "loss": 0.4104, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.9269406392694064, | |
| "grad_norm": 3.7278597354888916, | |
| "learning_rate": 5.47945205479452e-07, | |
| "loss": 0.4139, | |
| "step": 21500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 21900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1039565219447398e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |